The definitive therapy of localized prostate carcinoma is surgical in nature and other forms of therapy such as radiotherapy or endocrine ablation are reserved for recurrent or metastatic disease. Response to these latter forms of therapy remains unsatisfactory. Initial studies of prostate carcinoma cells line models of androgen- independent prostate cancer demonstrate a marked absence of AR expression. This finding prompted retrospective analysis that further suggested a link between the biological behavior of selected groups of patients with stage D prostate cancer and the levels of AR expression detected in initial biopsy specimens. The current proposal seeks to explore this work in four ways. First, we will attempt to confirm the relationship between AR expression in initial biopsy specimens and the subsequent clinical behavior of tumor specimens. Second, to investigate the molecular mechanisms controlling AR expression, we will characterize the proteins in prostate that bind to functionally important elements in the AR promoter. We have isolated four CDNA clones that bind to defined sequences in the AR promoter. We will isolate complete copies of these CDNAS and determine what role the proteins they encode play in controlling AR expression in the human prostate. Third, we will examine the molecular basis of the absence of AR expression in AR negative prostate cancer cell lines. Fourth, we will explore possible links between AR expression and cell growth.